Limits...
Genomic analysis of three African strains of Bacillus anthracis demonstrates that they are part of the clonal expansion of an exclusively pathogenic bacterium.

Rouli L, MBengue M, Robert C, Ndiaye M, La Scola B, Raoult D - New Microbes New Infect (2014)

Bottom Line: We found that the three African strains contained the pX01 and pX02 plasmids, the nonsense mutation in the plcR gene and the four known prophages.We validated the hypothesis that B. anthracis has a closed pan-genome and found that the three African strains carry the two plasmids associated with bacterial virulence.Moreover, thanks to the study of the core content single nucleotide polymorphisms, we can see that our three African strains diverged very recently from the other B. anthracis strains.

View Article: PubMed Central - PubMed

Affiliation: Aix Marseille Université, URMITE, UM63, CNRS 7278, IRD 198, Inserm 1095 Marseille, France.

ABSTRACT
Bacillus anthracis is the causative agent of anthrax and is classified as a 'Category A' biological weapon. Six complete genomes of B. anthracis (A0248, Ames, Ames Ancestor, CDC684, H0491, and Sterne) are currently available. In this report, we add three African strain genomes: Sen2Col2, Sen3 and Gmb1. To study the pan-genome of B. anthracis, we used bioinformatics tools, such as Cluster of Orthologous Groups, and performed phylogenetic analysis. We found that the three African strains contained the pX01 and pX02 plasmids, the nonsense mutation in the plcR gene and the four known prophages. These strains are most similar to the CDC684 strain and belong to the A cluster. We estimated that the B. anthracis pan-genome has 2893 core genes (99% of the genome size) and 85 accessory genes. We validated the hypothesis that B. anthracis has a closed pan-genome and found that the three African strains carry the two plasmids associated with bacterial virulence. The pan-genome nature of B. anthracis confirms its lack of exchange (similar to Clostridium tetani) and supports its exclusively pathogenic role, despite its survival in the environment. Moreover, thanks to the study of the core content single nucleotide polymorphisms, we can see that our three African strains diverged very recently from the other B. anthracis strains.

No MeSH data available.


Related in: MedlinePlus

Single nucleotide polymorphisms of the core content based tree. The percentage on the branchescorresponds to bootstraps. This is a PhyML tree with 100 bootstrap iterations. Colours correspond tothe species. Purple is Bacillus anthracis, pink is our three AfricanB. anthracis strains, blue is Bacillus cereus and brown isBacillus thuringiensis. (a) Numbers on branches correspond to bootstrap values. (b)Time tree based on the previous tree. Numbers on the branches are the divergence time, given inmillion years.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4265047&req=5

fig06: Single nucleotide polymorphisms of the core content based tree. The percentage on the branchescorresponds to bootstraps. This is a PhyML tree with 100 bootstrap iterations. Colours correspond tothe species. Purple is Bacillus anthracis, pink is our three AfricanB. anthracis strains, blue is Bacillus cereus and brown isBacillus thuringiensis. (a) Numbers on branches correspond to bootstrap values. (b)Time tree based on the previous tree. Numbers on the branches are the divergence time, given inmillion years.

Mentions: The pan-genome is composed of 2893 core genes, seven unique genes and 85 accessory genes. First,we considered the unique genes. We found five in Sterne (two not found on NCBI, one conservedhypothetical protein, an EmrB/QacA family drug resistance transporter and a zinc-bindingdehydrogenase), one in CDC684 (not found on NCBI) and one in H9401 (a yfeT DNA-bindingtranscriptional regulator), but all were false positives, as they were eventually identified inother B. anthracis strains available on the NCBI database. Next, we carefullyexamined the 85 accessory genes. The three African strains and CDC684 contained almost all of theaccessory genes. Half of the accessory genes were annotated as hypothetical proteins. Thehierarchical clustering again resulted in the same two groups found using COG and KEGG (one withAmes, Ames Ancestor and A0248 and the second with the other strains). Moreover, we calculated thecore/pan-genome ratio and found that the core genome represented 99% of the pan-genome(Table1), again indicating the high rate of conservationamong the nine strains. Finally, we studied the SNPs at the core genomic level. We foundapproximately 3800 SNPs between the nine strains, with 2911 in the core. Among the three Africanstrains, there are 1500 SNPs in the core and 1120 among the six reference strains. We can see thatthe three African strains cluster together. Moreover, we have a transition/transversion bias of 0.32between the nine strains. The very small rate of SNPs, low transition/transversion bias and veryhigh ratio of the core genome function to the pan-genome indicates thatB. anthracis evolves very slowly. To go further, we generated amaximum-likelihood tree based on SNPs of the core of the nine B. anthracisstrains, together with eight B. cereus, three Bacillusthuringiensis and, as an outgroup, one Bacillus cytotoxicus (Fig.6a) strain. This tree clearly showed that our three Africanstrains evolved differently from the other six strains of B. anthracis andthat is an ancestral difference. Moreover, we obtained good bootstraps, so we could be confidentabout this cluster. Finally, we generated a time tree (Fig.6b) to estimate the divergence time between each cluster. On Fig.6b, the number on branches are indicated in millions of years. For instance,between the African cluster and B. anthracis H9401, the estimated divergencetime is approximately 360 years. Moreover, the estimated divergence time between theB. anthracis cluster and B. cereus AH820 (n15 onFig.6b) is approximately 12 000 years. Forcomparison, in a work in 1999, Achtman et al. first estimated the divergencetime between Yersinia pestis and Yersinia pseudotuberculosis to bebetween 1500 and 20 000 years. Then, in 2004 32, they looked at different pathovars of Y. pestis. They found theestimated divergence time between an Orientalis strain (CO92) and a Medievalis strain (KIM) to beapproximately 6500 years.


Genomic analysis of three African strains of Bacillus anthracis demonstrates that they are part of the clonal expansion of an exclusively pathogenic bacterium.

Rouli L, MBengue M, Robert C, Ndiaye M, La Scola B, Raoult D - New Microbes New Infect (2014)

Single nucleotide polymorphisms of the core content based tree. The percentage on the branchescorresponds to bootstraps. This is a PhyML tree with 100 bootstrap iterations. Colours correspond tothe species. Purple is Bacillus anthracis, pink is our three AfricanB. anthracis strains, blue is Bacillus cereus and brown isBacillus thuringiensis. (a) Numbers on branches correspond to bootstrap values. (b)Time tree based on the previous tree. Numbers on the branches are the divergence time, given inmillion years.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4265047&req=5

fig06: Single nucleotide polymorphisms of the core content based tree. The percentage on the branchescorresponds to bootstraps. This is a PhyML tree with 100 bootstrap iterations. Colours correspond tothe species. Purple is Bacillus anthracis, pink is our three AfricanB. anthracis strains, blue is Bacillus cereus and brown isBacillus thuringiensis. (a) Numbers on branches correspond to bootstrap values. (b)Time tree based on the previous tree. Numbers on the branches are the divergence time, given inmillion years.
Mentions: The pan-genome is composed of 2893 core genes, seven unique genes and 85 accessory genes. First,we considered the unique genes. We found five in Sterne (two not found on NCBI, one conservedhypothetical protein, an EmrB/QacA family drug resistance transporter and a zinc-bindingdehydrogenase), one in CDC684 (not found on NCBI) and one in H9401 (a yfeT DNA-bindingtranscriptional regulator), but all were false positives, as they were eventually identified inother B. anthracis strains available on the NCBI database. Next, we carefullyexamined the 85 accessory genes. The three African strains and CDC684 contained almost all of theaccessory genes. Half of the accessory genes were annotated as hypothetical proteins. Thehierarchical clustering again resulted in the same two groups found using COG and KEGG (one withAmes, Ames Ancestor and A0248 and the second with the other strains). Moreover, we calculated thecore/pan-genome ratio and found that the core genome represented 99% of the pan-genome(Table1), again indicating the high rate of conservationamong the nine strains. Finally, we studied the SNPs at the core genomic level. We foundapproximately 3800 SNPs between the nine strains, with 2911 in the core. Among the three Africanstrains, there are 1500 SNPs in the core and 1120 among the six reference strains. We can see thatthe three African strains cluster together. Moreover, we have a transition/transversion bias of 0.32between the nine strains. The very small rate of SNPs, low transition/transversion bias and veryhigh ratio of the core genome function to the pan-genome indicates thatB. anthracis evolves very slowly. To go further, we generated amaximum-likelihood tree based on SNPs of the core of the nine B. anthracisstrains, together with eight B. cereus, three Bacillusthuringiensis and, as an outgroup, one Bacillus cytotoxicus (Fig.6a) strain. This tree clearly showed that our three Africanstrains evolved differently from the other six strains of B. anthracis andthat is an ancestral difference. Moreover, we obtained good bootstraps, so we could be confidentabout this cluster. Finally, we generated a time tree (Fig.6b) to estimate the divergence time between each cluster. On Fig.6b, the number on branches are indicated in millions of years. For instance,between the African cluster and B. anthracis H9401, the estimated divergencetime is approximately 360 years. Moreover, the estimated divergence time between theB. anthracis cluster and B. cereus AH820 (n15 onFig.6b) is approximately 12 000 years. Forcomparison, in a work in 1999, Achtman et al. first estimated the divergencetime between Yersinia pestis and Yersinia pseudotuberculosis to bebetween 1500 and 20 000 years. Then, in 2004 32, they looked at different pathovars of Y. pestis. They found theestimated divergence time between an Orientalis strain (CO92) and a Medievalis strain (KIM) to beapproximately 6500 years.

Bottom Line: We found that the three African strains contained the pX01 and pX02 plasmids, the nonsense mutation in the plcR gene and the four known prophages.We validated the hypothesis that B. anthracis has a closed pan-genome and found that the three African strains carry the two plasmids associated with bacterial virulence.Moreover, thanks to the study of the core content single nucleotide polymorphisms, we can see that our three African strains diverged very recently from the other B. anthracis strains.

View Article: PubMed Central - PubMed

Affiliation: Aix Marseille Université, URMITE, UM63, CNRS 7278, IRD 198, Inserm 1095 Marseille, France.

ABSTRACT
Bacillus anthracis is the causative agent of anthrax and is classified as a 'Category A' biological weapon. Six complete genomes of B. anthracis (A0248, Ames, Ames Ancestor, CDC684, H0491, and Sterne) are currently available. In this report, we add three African strain genomes: Sen2Col2, Sen3 and Gmb1. To study the pan-genome of B. anthracis, we used bioinformatics tools, such as Cluster of Orthologous Groups, and performed phylogenetic analysis. We found that the three African strains contained the pX01 and pX02 plasmids, the nonsense mutation in the plcR gene and the four known prophages. These strains are most similar to the CDC684 strain and belong to the A cluster. We estimated that the B. anthracis pan-genome has 2893 core genes (99% of the genome size) and 85 accessory genes. We validated the hypothesis that B. anthracis has a closed pan-genome and found that the three African strains carry the two plasmids associated with bacterial virulence. The pan-genome nature of B. anthracis confirms its lack of exchange (similar to Clostridium tetani) and supports its exclusively pathogenic role, despite its survival in the environment. Moreover, thanks to the study of the core content single nucleotide polymorphisms, we can see that our three African strains diverged very recently from the other B. anthracis strains.

No MeSH data available.


Related in: MedlinePlus